512 S. R. DETWILER 
Reviewing these few cases as a whole, we see that they showed 
no consistent defects, yet it is hardly to be expected that the 
removal of either the anterior or the posterior half of the limb 
dise would bring about any marked deficiencies in the girdle, since 
only a very small portion of the girdle rudiment can be in- 
volved in either case. Even after the removal of the entire limb 
disc, the suprascapula, procoracoid, and the coracoid develop in 
their normal position, while only the scapula and the glenoid 
portion of the girdle are wanting. It is quite possible that the 
excision of larger anterior or posterior areas would produce more 
pronounced and definite defects. 
B. Transplantation experiments: 
The development of a limb in the absence of either the dorsal 
or the ventral half of the girdle suggested the possibility of its 
development in the absence of the entire girdle. Experiments 
were therefore made to test this possibility. 
1. Transplantation of small areas of cells from the dorsal half of 
the limb disc. The area of mesoderm transplanted in these ex- 
periments is shown in text figure 1 by the smaller of the two 
circles. In five cases out of fifteen such transplants the tissue 
was resorbed. The other ten developed into limbs which showed 
more or less abnormality, consisting chiefly of reduplicated hands. 
Most of the limbs were atrophic and all lacked function. It was 
found when these larvae were sectioned twenty days after the 
operation that only a very fragmentary piece of the girdle was 
present, which from its relation to the humerus must be regarded 
as a rudimentary coracoid (fig. 14, cor. and fig. 31, cor). This 
was connected with the humerus by a very short supracora- 
coideus muscle. 
The almost complete absence of any girdle in these cases dem- 
onstrates that the region from which the transplanted tissue was 
taken is relatively free from cells which have the potency to 
form a girdle. When a larger area is transplanted, such as is 
indicated by the larger of the two circles in text figure 1, a con- 
